Voice coding format conversion adaptation method and system for integrated networking communication

Abstract

The invention discloses a voice coding format conversion adaptation method and system for integrated networking communication, and the method specifically comprises the steps: building an integrated networking communication system, and setting a voice coding format adaptation module in the integrated networking communication system; a source user selects a target user to initiate a voice service application, and a voice coding format adaptation module establishes a voice communication link; the voice coding format adaptation module receives voice data sent by a source user, converts the voice data into voice data in a voice coding format used by a target user, and sends the voice data to the target user; in the voice communication process, an MPI + OpenMP hybrid model is adopted to realize multi-channel parallel processing of coding format conversion, and a multi-level queue scheduling technology is utilized to perform scheduling control on different levels of voice service transmission requests; and after the voice communication service is finished, the voice coding format adaptation module removes the voice communication link. According to the invention, conversion and adaptation among multiple paths of parallel different voice coding formats are realized, and support is provided for realizing efficient, reliable and energy-saving voice information transmission.

Description

Technical Field

[0001] This invention relates to the field of integrated networking communication technology, and in particular to a voice coding format conversion and adaptation method and system for integrated networking communication. Background Technology

[0002] During the development of voice coding standards, various different voice coding standards have emerged. Consequently, there are many different voice coding methods used in communication equipment, each catering to different user voice communication needs. For example, the most widely used voice coding format in fixed-line telephone networks is G.711, while enterprise IP telephony systems (VoIP) commonly use G.729. G.722 is becoming increasingly popular in some modern IP telephony and video conferencing systems. Common voice coding formats used in satellite communication include AMR / AMR-WB, CVSD, and G.729. Shortwave communication, due to bandwidth limitations, uses special vocoders for high-compression voice coding, primarily employing Hybrid Excitation Linear Predictive Coding (MELP) and Linear Predictive Coding (LCP). Different types of users cannot communicate with each other because their communication devices use different voice coding methods. However, in practical applications, these users often need to communicate with each other. For example, if a user needs to communicate with both landline users and satellite or shortwave users, then he / she would need to have a landline, a satellite communication device, and a shortwave communication device to communicate with three different types of users. This is inconvenient and costly.

[0003] To achieve efficient, reliable, and seamless information transmission and sharing, integrated network communication combines various communication methods such as wired networks, wireless LANs, satellite communication, shortwave communication, and 4G / 5G cellular communication into a unified network for collaborative operation. This overcomes the limitations of a single network and meets diverse communication needs in complex environments. In integrated network voice communication, it is even more crucial to address the interoperability issues between communication devices using different voice coding formats. This resolves the interconnection and interoperability of voice services between various communication devices, enabling mutual conversion and adaptation between multiple voice coding formats. Ultimately, this allows users of communication devices using different voice coding formats to communicate with each other. Summary of the Invention

[0004] The purpose of this invention is to provide a comprehensive network communication voice coding format conversion and adaptation method and system that can solve the problem of incompatibility due to different voice coding formats of different communication devices, save communication costs, and achieve efficient, reliable, and seamless voice information transmission and sharing.

[0005] The technical solution for achieving the objective of this invention is: a voice coding format conversion and adaptation method for integrated network communication, comprising the following steps:

[0006] Step 1: Set up a voice encoding format adaptation module in the integrated network communication system. Various user terminal devices access the integrated network communication system through their respective access nodes. The user terminal devices accessing the system use the device number as a unique identifier.

[0007] Step 2: When conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction.

[0008] Step 3: Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user. In this process, when multiple voice services are running simultaneously in the integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to realize the parallel processing of encoding format conversion. Furthermore, in the process of integrated network voice communication, multi-level queue scheduling technology is used to schedule and control the transmission requests of different levels of voice services according to the priority settings of different users and different tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

[0009] Step 4: After the voice communication service ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

[0010] Furthermore, in step 2, when conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction, as detailed below:

[0011] Step 2.1: The voice encoding format adaptation module receives the voice call command information issued by the source user;

[0012] Step 2.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol and extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format;

[0013] Step 2.3: The voice encoding format adaptation module analyzes the target user's network status information. If the target user is on the network, it establishes a voice link and maintains the voice communication status, and sends a call success receipt. If the target user is not on the network, it sends a call failure receipt and maintains the existing voice link information.

[0014] Furthermore, in step 2, when the source user selects the destination user to initiate a voice service application, and the source user initiates a group call service for multiple users of a certain type, the voice encoding format adaptation module first receives the voice group setting instruction, extracts the group member information, and maintains the group information locally according to the group result information; after receiving the voice group call application instruction, it extracts the key parameter information, including the source user information, group number, and voice link number, establishes the group call link, and feeds back the group call application result to the network management center of the integrated network communication system.

[0015] Furthermore, the voice data described in step 3 is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user, as follows:

[0016] Step 3.1: Define the data transmission protocol. Each packet of voice data information includes the source user type, source device number, source encoding format, destination device number, destination user type, and voice data information.

[0017] Step 3.2: Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module parses the voice data packet content according to the agreed protocol and extracts the source voice encoding format, destination user type, destination device number, and voice data content.

[0018] Step 3.3: The voice encoding format adaptation module queries the voice encoding formats supported by the target user device based on the target user type or target device number, and performs voice encoding format conversion judgment. If the target user and the source user are the same type of user, there is no need to convert the voice data encoding format, and the voice data is directly forwarded to the target user according to the transmission protocol; if the target user and the source user are different types of users, it obtains whether the target user supports one voice encoding format or multiple voice encoding formats.

[0019] Step 3.4: If the target user supports a voice encoding format, first decode the source voice data into 16-bit PCM data; then re-encode the decoded PCM data into the voice encoding format used by the target user; next, according to the transmission protocol, encapsulate the voice data converted from the encoding format along with other necessary information into a data packet; finally, obtain the voice link connectivity status and send the encapsulated data packet to the target user.

[0020] Step 3.5: If the target user supports multiple voice encoding formats, an adaptive target encoding format selection method is adopted. First, historical data of voice communication between users is extracted to record the voice encoding formats commonly used by various users and the network conditions when using a certain encoding format, including time, user location, and network status, thus forming a knowledge base. Then, machine learning methods are used to select a target voice encoding format to complete the encoding format conversion and distribution of voice information. When the target user appears for the first time, since there is no historical data, the voice encoding formats supported by various user communication devices that are pre-stored are queried, and the first supported encoding format is selected for conversion.

[0021] Step 3.6: When multiple voice services are running simultaneously in an integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to achieve multi-channel parallel processing of encoding format conversion.

[0022] Step 3.7: During voice communication, the voice encoding format adaptation module uses multi-level queue scheduling technology to schedule and control voice service transmission requests of different levels according to the priority settings of different users and tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

[0023] Furthermore, in step 3.4, the decoded PCM data is re-encoded into the speech encoding format used by the target user. Since different speech encoding formats have different sampling rates, a sampling rate conversion is required before encoding the PCM data into the target encoding format. The sampling rate of the 16-bit PCM data obtained after decoding is converted into the sampling rate required by the target encoding format.

[0024] Furthermore, the sampling rate conversion involves the following specific steps:

[0025] The input signal bit x is set with a sampling rate of fs_in, and the target output signal has a sampling rate of fs_out. During sampling rate conversion, the conversion ratio R = fs_out / fs_in is first calculated. When the sampling rate of the target encoding format signal is the same as the PCM data sampling rate, R = 1, and no sampling rate conversion is needed; the signal is directly output to the encoding module for encoding. When R > 1, upsampling is required; when R < 1, downsampling is required.

[0026] There are three scenarios for sampling rate conversion: integer multiple upsampling, integer multiple downsampling, and fractional multiple upsampling or downsampling. Integer multiple upsampling is achieved through interpolation filtering, integer multiple downsampling is achieved by first applying anti-aliasing filtering and then decimating, and fractional multiple sampling rate conversion (R=L / M) is achieved by first upsampling the signal by a factor of L and then downsampling it by a factor of M, where L and M are positive integers. Therefore, all three sampling rate conversion scenarios are unified into a fractional form of L / M, with integer multiple conversion being a special case of fractional multiple conversion. For integer multiple upsampling, L=R, M=1; for integer multiple downsampling, L=1, M=1 / R. A unified filter design is used to efficiently implement this through a multiphase filter structure. The optimal implementation method is selected for each of the three scenarios: a multiphase interpolator is implemented for integer multiple upsampling, a multiphase decimator is implemented for integer multiple downsampling, and a multiphase fractional converter is implemented for fractional conversion.

[0027] Furthermore, the multi-parallel processing of encoding format conversion using the "MPI+OpenMP" hybrid programming model described in step 3.6 is as follows:

[0028] MPI creates multiple independent processes during computation, and these processes communicate with each other through message passing. Each process has its own independent memory space. OpenMP, on the other hand, breaks down tasks into multiple threads and executes these threads simultaneously. By combining MPI with OpenMP, multiple cores on a single node can be managed by a single MPI, while the nodes communicate internally through OpenMP threads that share memory.

[0029] Furthermore, the voice encoding format adaptation module described in step 3.7 utilizes multi-level queue scheduling technology to schedule and control voice service transmission requests of different levels based on the priority settings of different users and tasks, prioritizing the voice transmission needs and real-time performance of high-priority or key users, as detailed below:

[0030] Four customized queues are set up for the voice communication relationships of the same user, corresponding to four priorities: normal, urgent, express, and super urgent. By maintaining priority strategies for different users and tasks, and through user selection, voice services of different priorities are cached in queues of different priorities. When a voice transmission task in a high-priority queue is established first, the high-priority queue is guaranteed to complete the transmission first, and the low-priority queue task enters a waiting state after it is established. When a voice transmission task in a high-priority queue is established later, the original low-priority queue voice transmission task is interrupted according to the scheduling strategy, and the high-priority voice transmission begins directly, while the low-priority queue task enters a waiting state.

[0031] Furthermore, after the voice communication service described in step 4 ends, the voice encoding format adaptation module dismantles the voice communication link according to the received voice service termination instruction, as follows:

[0032] Step 4.1: The voice encoding format adaptation module receives the voice call end instruction information;

[0033] Step 4.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol, extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format, finds the corresponding link in the existing links and deletes it;

[0034] Step 4.3: The voice encoding format adaptation module updates the voice link status information.

[0035] A voice coding format conversion and adaptation system for integrated network communication is disclosed. This system implements the aforementioned voice coding format conversion and adaptation method for integrated network communication. The system includes a voice coding format adaptation module and a user terminal device. The user terminal device includes a source user and a destination user, as detailed below:

[0036] A voice coding format adaptation module is set in the integrated network communication system. User terminal devices access the integrated network communication system through their respective access nodes. The user terminal devices accessing the system use the device number as a unique identifier.

[0037] When conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction.

[0038] Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user.

[0039] When multiple voice services are running simultaneously in an integrated network communication system, a hybrid programming model of "MPI+OpenMP" is used to achieve multi-parallel processing of encoding format conversion.

[0040] In the process of integrated network communication, multi-level queue scheduling technology is used to schedule and control voice service transmission requests of different levels according to the priority settings of different users and different tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

[0041] After the voice communication service ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

[0042] Compared with the prior art, the present invention has the following significant advantages: (1) It realizes the mutual conversion and adaptation between different voice coding formats in multi-channel parallel communication, and solves the problem that various users cannot communicate with each other because the voice coding formats of the communication devices they use are different when conducting integrated network voice communication; (2) It saves communication costs, improves the communication efficiency between various users when conducting integrated network voice communication, and provides support for the efficient, reliable and seamless transmission and sharing of voice information in integrated network communication. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating a voice encoding format conversion and adaptation method for integrated network communication according to the present invention.

[0044] Figure 2 This is a schematic diagram of the integrated networking communication system in this invention.

[0045] Figure 3 This is a schematic diagram of the process for establishing a voice communication link in this invention.

[0046] Figure 4 This is a schematic diagram of the process for establishing a group call link in this invention.

[0047] Figure 5 This is a schematic diagram illustrating the principle of different speech encoding format conversion in this invention.

[0048] Figure 6 This is a schematic diagram of the speech encoding format conversion process in this invention.

[0049] Figure 7 This is a schematic diagram of the sampling rate conversion process in this invention.

[0050] Figure 8 This is a schematic diagram of the process of disconnecting the voice communication link in this invention. Detailed Implementation

[0051] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0052] This invention provides a voice coding format conversion and adaptation method and system for integrated network communication, the working principle of which is illustrated in the diagram below. Figure 1As shown, in integrated networking, wired networks, satellite communications, shortwave communications, and 5G communications access the integrated networking communication system through their respective access nodes. User terminal devices accessing the system use address codes or device numbers as unique identifiers, collectively referred to as device numbers. The voice encoding format adaptation module is a functional module within the processing center of the integrated networking communication system. When conducting voice services, the call initiator (hereinafter referred to as the source user) selects the destination user to initiate a voice service request. The voice encoding format adaptation module establishes a voice communication link based on the received voice service request instruction, and then the source user and destination user can conduct voice communication. Because various wired users, satellite terminal users, shortwave mobile users, 5G users, etc., use different voice encoding formats, during voice communication, voice data is transmitted through the source user's "access node" to the voice encoding format adaptation module of the integrated networking communication system. The voice encoding format adaptation module converts the received voice data in the source user's voice encoding format (hereinafter referred to as the source encoding format) into the voice encoding format used by the destination user (hereinafter referred to as the destination encoding format), and then sends it to the destination user through the destination user's "access node." After the voice communication service ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

[0053] like Figure 1 As shown, the present invention provides a voice coding format conversion and adaptation method for integrated network communication, comprising the following steps:

[0054] Step 1: Configure a voice coding format adaptation module in the integrated network communication system. User terminal devices, such as wired networks, satellite communications, shortwave communications, and 5G communications, access the integrated network communication system through their respective access nodes. Each user terminal device uses its device number as a unique identifier when accessing the system. Figure 2 As shown;

[0055] Step 2: When conducting voice services, the source user selects the destination user to initiate a voice service request. The voice encoding format adaptation module establishes a voice communication link based on the received voice service request instruction, such as... Figure 3 As shown, the details are as follows:

[0056] Step 2.1: The voice encoding format adaptation module receives the voice call command information issued by the source user;

[0057] Step 2.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol and extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format;

[0058] Step 2.3: The voice encoding format adaptation module analyzes the target user's network status information. If the target user is on the network, it establishes a voice link and maintains the voice communication status, and sends a call success receipt. If the target user is not on the network, it sends a call failure receipt and maintains the existing voice link information.

[0059] As a specific example, sometimes it is necessary to initiate a group call for multiple users of a certain type. When voice group call services are required, such as... Figure 4 As shown, when a source user initiates a group call service for multiple users of a certain type, the voice encoding format adaptation module first receives the voice group setting instruction, extracts the group member information, and maintains the group information locally according to the group result information. When it receives the voice group call application instruction, it extracts the key parameter information, including the source user information, group number, and voice link number, establishes the group call link, and feeds back the group call application result to the network management center of the integrated network communication system.

[0060] Step 3: Voice data is transmitted through the source user's access node to the voice encoding format adaptation module of the integrated network communication system. The voice encoding format adaptation module converts the received voice data in the source user's voice encoding format into voice data in the destination user's voice encoding format, and then sends it to the destination user through the destination user's access node. During this process, when multiple voice services are running simultaneously in the integrated network communication system, a hybrid programming model of "MPI+OpenMP" is used to achieve parallel processing of the encoding format conversion. Furthermore, during the integrated network voice communication process, multi-level queue scheduling technology is used to schedule and control the transmission requests of different levels of voice services based on the priority settings of different users and tasks, prioritizing the voice transmission needs and real-time performance of high-priority or key users. Details are as follows:

[0061] Step 3.1: Define the data transmission protocol. Each packet of voice data information includes the source user type, source device number, source encoding format, destination device number, destination user type, and voice data information.

[0062] Step 3.2: Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module parses the voice data packet content according to the agreed protocol and extracts the source voice encoding format, destination user type, destination device number, and voice data content. Since the voice encoding formats supported by the communication devices used by various users are not the same, the voice encoding formats supported by the communication devices of various users are stored in advance.

[0063] Step 3.3: After the voice encoding format adaptation module parses the received voice data packet, it can query the voice encoding formats supported by the destination user device based on the destination user type or destination webbing number. The voice encoding format adaptation module performs a voice encoding format conversion judgment based on the destination user type or destination webbing number. If the destination user and the source user are the same type of user, there is no need to convert the voice data encoding format, and the voice data is directly forwarded to the destination user according to the transmission protocol. If the destination user and the source user are different types of users, it determines whether the destination user supports one or multiple voice encoding formats.

[0064] Step 3.4: If the destination user supports a certain voice encoding format, first decode the source voice data into 16-bit PCM data; then re-encode the decoded PCM data into the voice encoding format used by the destination user; next, according to the transmission protocol, encapsulate the encoded voice data along with other necessary information into a data packet; finally, obtain the voice link connectivity status and send the encapsulated data packet to the destination user, such as... Figure 5 As shown in the figure, only some commonly used speech coding formats are listed, such as G.711, G.729, CVSD, MELP, LPC, Opus, AMR-WB, etc. In practical applications, speech data in other coding formats can also be processed using this method.

[0065] Step 3.5: If the target user supports multiple voice encoding formats, an adaptive target encoding format selection method is adopted. First, historical data of voice communication between users is extracted to record the voice encoding formats commonly used by various users and the network conditions when using a certain encoding format, including time, user location, and network status, thus forming a knowledge base. Then, machine learning methods are used to select a target voice encoding format to complete the encoding format conversion and distribution of voice information. When the target user appears for the first time, since there is no historical data, the voice encoding formats supported by various user communication devices that are pre-stored are queried, and the first supported encoding format is selected for conversion.

[0066] As a specific example, the speech coding format adaptation module's adaptive speech coding format conversion and data distribution process described in steps 3.1 to 3.5 is as follows: Figure 6 As shown.

[0067] Step 3.6: When multiple voice services are running simultaneously in an integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to achieve multi-channel parallel processing of encoding format conversion.

[0068] Step 3.7: During voice communication, the voice encoding format adaptation module uses multi-level queue scheduling technology to schedule and control voice service transmission requests of different levels according to the priority settings of different users and tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

[0069] As a specific example, step 3.4 involves re-encoding the decoded PCM data into the speech encoding format used by the target user. Since different speech encoding formats have different sampling rates, a sampling rate conversion is required before encoding the PCM data into the target encoding format. This involves converting the sampling rate of the decoded 16-bit PCM data to the sampling rate required by the target encoding format. Figure 7 As shown, the specific steps are as follows:

[0070] The input signal bit x is set with a sampling rate of fs_in, and the target output signal has a sampling rate of fs_out. During sampling rate conversion, the conversion ratio R = fs_out / fs_in is first calculated. When the sampling rate of the target encoding format signal is the same as the PCM data sampling rate, R = 1, and no sampling rate conversion is needed; the signal is directly output to the encoding module for encoding. When R > 1, upsampling is required; when R < 1, downsampling is required.

[0071] There are three scenarios for sampling rate conversion: integer multiple upsampling, integer multiple downsampling, and fractional multiple upsampling or downsampling. Integer multiple upsampling can be achieved through interpolation filtering, and integer multiple downsampling can be achieved by first using anti-aliasing filtering and then decimation. Fractional sampling rate conversion with a conversion rate of R=L / M can be achieved by first upsampling the signal by a factor of L and then downsampling it by a factor of M, where L and M are positive integers. Therefore, all three sampling rate conversion scenarios are unified into the fractional form of L / M. Integer multiple conversion is simply a special case of fractional conversion. For integer multiple upsampling, L=R, M=1; for integer multiple downsampling, L=1, M=1 / R. A unified filter design is used to efficiently implement this through a multiphase filter structure. The optimal implementation method is selected for each of the three scenarios: a multiphase interpolator is implemented for integer multiple upsampling, a multiphase decimator is implemented for integer multiple downsampling, and a multiphase fractional converter is implemented for fractional conversion.

[0072] As a specific example, the multi-parallel processing of encoding format conversion using the "MPI+OpenMP" hybrid programming model described in step 3.6 is as follows:

[0073] In integrated network communication, multiple voice services often need to run simultaneously. To improve processing efficiency, this invention employs a hybrid programming model of "MPI + OpenMP" to achieve multi-channel parallel processing of encoding format conversion. MPI creates multiple independent processes during computation, communicating with each other via message passing. Each process has its own independent memory space. When the number of processes is very large, the inter-process communication overhead is significant, easily causing congestion. OpenMP, on the other hand, decomposes tasks into multiple threads, executing multiple threads simultaneously. Therefore, by combining MPI with OpenMP, multiple cores on a single node are managed by a single MPI, while internal communication within the node is achieved through shared memory via OpenMP threads. This significantly reduces the number of MPI processes and substantially lowers communication pressure.

[0074] As a specific example, in step 3.7, during voice communication, the voice encoding format adaptation module utilizes multi-level queue scheduling technology. Based on the priority settings of different users and tasks, it schedules and controls voice service transmission requests of different levels, prioritizing the voice transmission needs and real-time performance of high-priority or key users. Specifically:

[0075] Four customized queues are set up for the voice communication relationships of the same user, corresponding to four priorities: normal, urgent, express, and super urgent. By maintaining priority strategies for different users and tasks, and through user selection, voice services of different priorities are cached in queues of different priorities. When a voice transmission task in a high-priority queue is established first, the high-priority queue is prioritized for completion, and the transmission task in a low-priority queue enters a waiting state after its establishment. When a voice transmission task in a high-priority queue is established later, the existing low-priority queue voice transmission task can be interrupted according to the scheduling strategy, and the high-priority voice transmission can start directly, while the low-priority queue task enters a waiting state. Through multi-level queue scheduling technology, network resources can be rationally allocated and resource utilization efficiency can be improved.

[0076] Step 4: After the voice communication service ends, the voice encoding format adaptation module dismantles the voice communication link according to the received voice service termination instruction, such as... Figure 8 As shown, the details are as follows:

[0077] Step 4.1: The voice encoding format adaptation module receives the voice call end instruction information;

[0078] Step 4.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol, extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format, finds the corresponding link in the existing links and deletes it;

[0079] Step 4.3: The voice encoding format adaptation module updates the voice link status information.

[0080] This invention also provides a voice coding format conversion and adaptation system for integrated network communication. This system implements the aforementioned voice coding format conversion and adaptation method for integrated network communication. The system includes a voice coding format adaptation module and a user terminal device. The user terminal device includes a source user and a destination user, as detailed below:

[0081] A voice coding format adaptation module is set in the integrated network communication system. User terminal devices access the integrated network communication system through their respective access nodes. The user terminal devices accessing the system use the device number as a unique identifier.

[0082] When conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction.

[0083] Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user.

[0084] When multiple voice services are running simultaneously in an integrated network communication system, a hybrid programming model of "MPI+OpenMP" is used to achieve multi-parallel processing of encoding format conversion.

[0085] In the process of integrated network communication, multi-level queue scheduling technology is used to schedule and control voice service transmission requests of different levels according to the priority settings of different users and different tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

[0086] After the voice communication ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

[0087] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A voice coding format conversion and adaptation method for integrated network communication, characterized in that, Includes the following steps: Step 1: Set up a voice encoding format adaptation module in the integrated network communication system. Various user terminal devices access the integrated network communication system through their respective access nodes. The user terminal devices accessing the system use the device number as a unique identifier. Step 2: When conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction. Step 3: The voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user. In this process, when multiple voice services are running simultaneously in the integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to achieve multi-channel parallel processing of encoding format conversion. Furthermore, in the process of integrated network voice communication, multi-level queue scheduling technology is used to schedule and control the transmission requests of different levels of voice services according to the priority settings of different users and different tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users. Step 4: After the voice communication service ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

2. The voice coding format conversion and adaptation method for integrated network communication according to claim 1, characterized in that, In step 2, when conducting voice services, the source user selects the destination user to initiate a voice service request. The voice encoding format adaptation module establishes a voice communication link based on the received voice service request instruction, as detailed below: Step 2.1: The voice encoding format adaptation module receives the voice call command information issued by the source user; Step 2.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol and extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format; Step 2.3: The voice encoding format adaptation module analyzes the target user's network status information. If the target user is on the network, it establishes a voice link and maintains the voice communication status, and sends a call success receipt. If the target user is not on the network, a call failure receipt will be sent out, and existing voice link information will be maintained.

3. The voice coding format conversion and adaptation method for integrated network communication according to claim 1, characterized in that, In step 2, when the source user selects the destination user to initiate a voice service application, and the source user initiates a group call service for multiple users of a certain type, the voice encoding format adaptation module first receives the voice group setting instruction, extracts the group member information, and maintains the group information locally according to the group result information; after receiving the voice group call application instruction, it extracts the key parameter information, including the source user information, group number, and voice link number, establishes the group call link, and feeds back the group call application result to the network management center of the integrated network communication system.

4. The voice coding format conversion and adaptation method for integrated network communication according to claim 1, characterized in that, The voice data mentioned in step 3 is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user, as follows: Step 3.1: Define the data transmission protocol. Each packet of voice data information includes the source user type, source device number, source encoding format, destination device number, destination user type, and voice data information. Step 3.2: Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module parses the voice data packet content according to the agreed protocol and extracts the source voice encoding format, destination user type, destination device number, and voice data content. Step 3.3: The voice encoding format adaptation module queries the voice encoding formats supported by the target user device based on the target user type or target device number, and performs voice encoding format conversion judgment. If the target user and the source user are the same type of user, then there is no need to convert the encoding format of the voice data, and the voice data is directly forwarded to the target user according to the transmission protocol. If the target user and the source user are different types of users, then determine whether the target user supports one or more voice encoding formats. Step 3.4: If the target user supports a certain voice encoding format, first decode the source voice data into 16-bit PCM data; Then, the decoded PCM data is re-encoded into the voice encoding format used by the target user; next, according to the transmission protocol, the voice data converted from the encoding format is encapsulated into a data packet along with other necessary information; finally, the voice link connection status is obtained, and the encapsulated data packet is sent to the target user. Step 3.5: If the target user supports multiple voice encoding formats, an adaptive target encoding format selection method is adopted. First, historical data of voice communication between users is extracted to record the voice encoding formats commonly used by various users and the network conditions when using a certain encoding format, including time, user location, and network status, thus forming a knowledge base. Then, machine learning methods are used to select a target voice encoding format to complete the encoding format conversion and distribution of voice information. When the target user appears for the first time, since there is no historical data, the voice encoding formats supported by various user communication devices that are pre-stored are queried, and the first supported encoding format is selected for conversion. Step 3.6: When multiple voice services are running simultaneously in an integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to achieve multi-channel parallel processing of encoding format conversion. Step 3.7: During voice communication, the voice encoding format adaptation module uses multi-level queue scheduling technology to schedule and control voice service transmission requests of different levels according to the priority settings of different users and tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users.

5. The voice coding format conversion and adaptation method for integrated network communication according to claim 4, characterized in that, Step 3.4 describes re-encoding the decoded PCM data into the speech encoding format used by the target user. Since different speech encoding formats have different sampling rates, a sampling rate conversion is required before encoding the PCM data into the target encoding format. The sampling rate of the 16-bit PCM data obtained after decoding is converted into the sampling rate required by the target encoding format.

6. The voice coding format conversion and adaptation method for integrated network communication according to claim 5, characterized in that, The sampling rate conversion process is as follows: The input signal bit x is set with a sampling rate of fs_in, and the target output signal has a sampling rate of fs_out. During sampling rate conversion, the conversion ratio R = fs_out / fs_in is first calculated. When the sampling rate of the target encoding format signal is the same as the PCM data sampling rate, R = 1, and no sampling rate conversion is needed; the signal is directly output to the encoding module for encoding. When R > 1, upsampling is required; when R < 1, downsampling is required. There are three scenarios for sampling rate conversion: integer multiple upsampling, integer multiple downsampling, and fractional multiple upsampling or downsampling. Integer multiple upsampling is achieved through interpolation filtering, integer multiple downsampling is achieved by first applying anti-aliasing filtering and then decimating, and fractional multiple sampling rate conversion (R=L / M) is achieved by first upsampling the signal by a factor of L and then downsampling it by a factor of M, where L and M are positive integers. Therefore, all three sampling rate conversion scenarios are unified into a fractional form of L / M, with integer multiple conversion being a special case of fractional multiple conversion. For integer multiple upsampling, L=R, M=1; for integer multiple downsampling, L=1, M=1 / R. A unified filter design is used to efficiently implement this through a multiphase filter structure. The optimal implementation method is selected for each of the three scenarios: a multiphase interpolator is implemented for integer multiple upsampling, a multiphase decimator is implemented for integer multiple downsampling, and a multiphase fractional converter is implemented for fractional conversion.

7. The voice coding format conversion and adaptation method for integrated network communication according to claim 4, characterized in that, Step 3.6 describes the implementation of multi-channel parallel processing for encoding format conversion using the "MPI+OpenMP" hybrid programming model, as follows: MPI creates multiple independent processes during computation, and these processes communicate with each other through message passing. Each process has its own independent memory space. OpenMP, on the other hand, breaks down tasks into multiple threads and executes these threads simultaneously. By combining MPI with OpenMP, multiple cores on a single node can be managed by a single MPI, while the nodes communicate internally through OpenMP threads that share memory.

8. The voice coding format conversion and adaptation method for integrated network communication according to claim 4, characterized in that, The voice encoding format adaptation module described in step 3.7 utilizes multi-level queue scheduling technology to schedule and control voice service transmission requests of different levels based on the priority settings of different users and tasks. It prioritizes the voice transmission needs and real-time performance of high-priority or key users, as detailed below: Four customized queues are set up for the voice communication relationships of the same user, corresponding to four priorities: normal, urgent, express, and super urgent. By maintaining priority strategies for different users and tasks, and through user selection, voice services of different priorities are cached in queues of different priorities. When a voice transmission task in a high-priority queue is established first, the high-priority queue is guaranteed to complete the transmission first, and the low-priority queue task enters a waiting state after it is established. When a voice transmission task in a high-priority queue is established later, the original low-priority queue voice transmission task is interrupted according to the scheduling strategy, and the high-priority voice transmission begins directly, while the low-priority queue task enters a waiting state.

9. The voice coding format conversion and adaptation method for integrated network communication according to claim 1, characterized in that, After the voice communication service described in step 4 ends, the voice encoding format adaptation module dismantles the voice communication link according to the received voice service termination instruction, as follows: Step 4.1: The voice encoding format adaptation module receives the voice call end instruction information; Step 4.2: The voice encoding format adaptation module parses the instruction content according to the agreed protocol, extracts key parameter information of the source user and the destination user, including device number, voice link number, and source user voice encoding format, finds the corresponding link in the existing links and deletes it; Step 4.3: The voice encoding format adaptation module updates the voice link status information.

10. A voice coding format conversion and adaptation system for integrated network communication, characterized in that, This system is used to implement the voice coding format conversion and adaptation method for integrated network communication as described in any one of claims 1 to 9. The system includes a voice coding format adaptation module and a user terminal device. The user terminal device includes a source user and a destination user, as detailed below: A voice coding format adaptation module is set in the integrated network communication system. User terminal devices access the integrated network communication system through their respective access nodes. The user terminal devices accessing the system use the device number as a unique identifier. When conducting voice services, the source user selects the destination user to initiate a voice service application, and the voice encoding format adaptation module establishes a voice communication link according to the received voice service application instruction. Voice data is transmitted to the voice encoding format adaptation module of the integrated network communication system through the access node of the source user. The voice encoding format adaptation module converts the received voice data in the voice encoding format of the source user into voice data in the voice encoding format used by the destination user, and sends it to the destination user through the access node of the destination user. When multiple voice services are running simultaneously in an integrated network communication system, the "MPI+OpenMP" hybrid programming model is used to achieve multi-parallel processing of encoding format conversion. In the process of integrated network communication, multi-level queue scheduling technology is used to schedule and control voice service transmission requests of different levels according to the priority settings of different users and different tasks, so as to give priority to ensuring the voice transmission needs and real-time performance of high-priority users or key users. After the voice communication service ends, the voice encoding format adaptation module disconnects the voice communication link according to the received voice service termination instruction.

Images